4.6 Article

Projection and proximal point methods:: convergence results and counterexamples

Journal

NONLINEAR ANALYSIS-THEORY METHODS & APPLICATIONS
Volume 56, Issue 5, Pages 715-738

Publisher

PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.na.2003.10.010

Keywords

alternating projections; averaged projections; Hilbert space; nonexpansive; proximal point algorithm; weak convergence

Ask authors/readers for more resources

Recently, Hundal has constructed a hyperplane H, a cone K, and a starting point y(0) in l(2) such that the sequence of alternating projections ((PKPH)(n)y(0))nepsilonN converges weakly to some point in H boolean AND K, but not in norm. We show how this construction results in a counterexample to norm convergence for iterates of averaged projections; hence, we give an affirmative answer to a question raised by Reich two decades ago. Furthermore, new counterexamples to norm convergence for iterates of firmly nonexpansive maps (a la Genel and Lindenstrauss) and for the proximal point algorithm (a la Guler) are provided. We also present a counterexample, along with some weak and norm convergence results, for the new framework of string-averaging projection methods introduced by Censor et at. Extensions to Banach spaces and the situation for the Hilbert ball are discussed as well. (C) 2003 Elsevier Ltd. All rights reserved.

Authors

I am an author on this paper
Click your name to claim this paper and add it to your profile.

Reviews

Primary Rating

4.6
Not enough ratings

Secondary Ratings

Novelty
-
Significance
-
Scientific rigor
-
Rate this paper

Recommended

No Data Available
No Data Available